Lumbosacral radicular pain is present in around 40% of cases of low back pain. It is usually caused by irritation and inflammation of the nerve root, and patients typically experience pain radiating from the back to the lower limb in the distribution of the affected spinal nerves. Epidural steroid injection is the most commonly used pain-relieving procedure in the world. However, the analgesic efficacy of epidural steroid injection appears to be modest and duration limited. Platelet rich plasma (PRP) is an emerging treatment option for chronic pain. It is currently used for treating musculoskeletal pain conditions such as osteoarthritis and tendinopathies. While epidural steroids reduce pain by reducing inflammation, PRP promotes the healing of nerve injury and reduces neuropathic pain. However, the effect of transforaminal epidural PRP versus epidural steroid specifically for lumbosacral radicular pain is unclear. In this study, a double blind, randomized controlled trial will be performed to compare the effect of transforaminal epidural PRP versus epidural steroid for pain relief in patients with lumbosacral radicular pain.

Procedures:

Preparation of PRP The handling of and preparation of PRP will be performed under sterile conditions in the operating theatre of Queen Mary Hospital, Hong Kong, China. The SmartPReP Platelet Concentrate System (Harvest Terumo BCT) will be used and delivers a concentration of platelets and growth factors that is roughly 4-6 times greater than that found in whole blood. The separation process produces a white blood cell (WBC) composition that contains 100% more mononuclear (lymphocytes and monocytes) cells while reducing the granulocyte cells by 60%. Two millilitres of anticoagulant citrate solution will be first added into the smaller chamber of a two-chamber APC (Autologous Platelet Concentrate, Harvest Terumo BCT) kit. Fifty-four milli-litres of autologous venous blood will be drawn under aseptic technique from the cephalic or basilic vein of each patient's upper limb and collected in a sterile 60 cc syringe that contains 6 ml of anticoagulant citrate solution. The total volume of 60ml consisting of blood sample and citrate solution will be transferred to the larger chamber (blood chamber) of the APC kit. The APC kit will be loaded in the SmartPRep2 centrifuge machine. Centrifuge spinning procedures will be followed by standard two-stage centrifugation spinning. The first cycle will take three minutes at 2500 ± 150 rpm and the second cycle will take nine minutes at 2300 ±140 rpm. After processing, the syringe with the spacer will be used to withdraw and discard the platelet poor plasma layer. Seven milli-litres of PRP will finally collected for injections. For patients in the ESI group, the blood will be discarded. These will be done after the patient is placed in the prone position to maintain patient blinding.

Epidural injection All epidural injections will be performed under aseptic techniques using fluoroscopic guidance. A specialist pain fellow accredited by the Hong Kong College of Anaesthesiologists will either conduct or supervise the performance of epidural steroid or PRP injection. Epidural injection will be performed under aseptic technique in the operating theatre. Standard monitoring will be applied and patients will receive light sedation using target-controlled infusion (TCI) of propofol with an effect site concentration of titrated between 0.5-1.5mcg/ml.

The lumbosacral level selected for injection will be based on symptoms and imaging findings. The image intensifier will be adjusted to obtain an oblique view. A single level, unilateral epidural injection will be performed. For the transforaminal epidural injection, a 22-gauge spinal needle will be inserted coaxially via a subpedicular approach to target the upper part of the foramen under fluoroscopic guidance. Correct needle position will be confirmed with appropriate contrast spread (1ml of contrast) in the anteroposterior, oblique, and lateral view. Spread of contrast delineating the nerve root together with proximal epidural spread will be accepted as indicating correct needle position. A single level injection will be performed because there are no clinical trials showing that multilevel injections is superior to single level injection, and a well-positioned injection usually spreads to multiple levels, thus achieving the effect of multi-level injection. 2ml of 1% lignocaine will be given for skin infiltration. Once correct position is confirmed, drug injectate will be administered. For the ESI group, 8mg of dexamethasone (4mg/ml) added to 1 ml of normal saline will be given (total 3ml volume). For the EPRP group, 3 ml of PRP will be injected. The total volume of injectate for both groups of patients is 3ml.

Blinding will be assessed immediately after epidural injection using the James blinding index and Bing blinding index.

Co-interventions Patients from both groups will be instructed to stop taking non-steroidal anti-inflammatory drugs (NSAIDs), corticosteroids starting (2 weeks before and 6 weeks post procedure), and any other drugs that may affect platelet function starting from 2 weeks before epidural injection. Apart from these medications, all patients will continue with their current analgesic drug therapy. Tramadol 50mg orally every 6 hours as required can be prescribed as rescue medication. For patients taking opioids (including tramadol), these can be increased by up to 20% above baseline. Patients will be instructed on how to titrate their analgesic medication based on response during their follow up by the pain physician. All patients will receive a standardized physiotherapy program after epidural injection (2 sessions). Patients in the study are not allowed to receive other co-interventions such as other interventional pain procedures or acupuncture. But they can exit the study at any time point during the trial.

Postoperative follow up and outcome assessment Outcome data will be collected by an investigator who will be blinded to patient allocation. Baseline data for these outcomes will be collected prior to epidural injection. Outcome assessment will be performed during patient follow up on postoperative day (POD) 1, and also at 1, 3, and 6 months after epidural injection.

Data on the following pain outcomes will be collected: average NRS pain score over the leg (average pain score over the past week), average NRS pain score over the back, worse pain score over the leg (over the past week), worse pain score over the back, and the presence of 50% or more reduction in leg pain.

Other outcomes that will be assessed include health-related quality of life, functional status, patient global impression of change scale (PGIC), analgesic consumption, and psychological status. The Oswestry disability index (ODI) will be used to assess function. This a 10-question survey used to assess function for individuals with low back and/or leg pain. Health related quality of life will be assessed using the Chinese version of the Short Form 36-Item Health Survey (SF-36). EQ5D will be also be used to measure health status and to enable cost-effectiveness analysis. Psychological status will be evaluated using the Beck Depression Scale. Complications and adverse effects related to the procedure will be assessed using both closed ended and open-ended questions on the same day after the procedure, on POD 1, and at 1, 3 and 6 months after injection. These will be assessed by a blinded investigator. A reduction in analgesic medication will be defined as cessation of non-opioid analgesic medication or a 20% reduction in opioid consumption.

Patients who have a positive response to the epidural injection will continue to be blinded and assessed at the next follow-up. A positive response to epidural steroid or PRP injection is defined as a 50% or more reduction in average leg pain and positive global perceived effect. Positive global perceived effect is defined as a positive response to the question 'my pain has improved/worsened/stayed the same since my last visit' and 'I am satisfied/not satisfied with the treatment I received and would recommend it to others. Patients will not be permitted to receive other interventional pain procedures while they are in the study. Patients can choose to exit the study at any time to pursue other interventions (eg surgery, interventional pain procedures) if they feel inadequate benefit with the allocated treatment. Unblinding will be done for patients who exit the study and they will not be followed up. Intention to treat analysis will be performed. Missing data points for pain scores and other outcomes will be imputed using the 'last observation carried forward' method, which is also similar to other studies evaluating epidural injection.

Other data collection The following data will also be collected during the study period

• Demographics
• Duration of hospital stay
• Pathological diagnosis (eg herniated disc, spinal stenosis)
• Any precipitating event for lumbosacral radicular pain (none, road traffic accident, fall, lifting heavy object, other)
• Level of epidural injection
• Duration of lumbosacral radicular pain
• Analgesic medication (including opioids)
• Adverse events and complications
• Other pain comorbidities

Inclusion Criteria:

• Age 18 years or above
• average radicular leg pain of ≥ 4 (numerical rating scale (NRS) 0-10) over the past week,
• Unilateral radicular leg pain
• symptoms and/or signs of lumbosacral radicular pain (such as lower leg pain or back/buttock pain radiating to lower limb or positive straight leg raising or lower limb numbness or altered lower limb sensation)
• pain duration more than 6 weeks
• Imaging findings (magnetic resonance imaging) of a pathologic condition correlating with symptoms or signs (herniated disc or annular tear or spinal stenosis)
• able to provide informed consent

Exclusion Criteria:

• patients require surgical intervention
• coagulopathy (platelet count less than 100,000/ml, international normalized ratio [INR] over 1.5).
• Using anticoagulant and/or antiplatelet medication (not including aspirin)
• received epidural steroid injection within the past 1 year
• systemic infection
• previous lumbar spine surgery
• allergy: contrast dye, steroids, PRP, local anaesthetic
• pregnancy
• physically unable to tolerate epidural injection
• uncontrolled psychiatric disorder (Beck depression score ≥ 21)
• active or a history of substance use disorder (including alcohol, opioid)
• known secondary gain (eg active litigation)
• patients with predominantly neurogenic claudication (pain over the calves, buttock, and/or thigh triggered by walking and/or prolonged standing in upright position).
• back pain greater than leg pain